Table 12 Primary outcomes of systematic review: effect estimates for all trials and for high-quality trials

All trials with extractable data for meta-analysis

 Linde (1997) [6]

89

0

Random effects

OR

2.45 (2.05–2.93)

 > 1

Yes

 Linde (1998) [7]

18

0

Fixed effects

RR

1.66 (1.20–2.28)

 > 1

Yes

 Linde (1998) [7]

18

0

Fixed effects

OR

2.62

 > 1

Yes

 Cucherat (2000) [8]

17^a

0^b

Not applicable

p = 0.000036

NA

Yes

 Mathie (2014) [10]

22

0^c

Random effects

OR

1.53 (1.22–1.91)

 > 1

Yes

 Mathie (2017) [11]

54

0^c

Random effects

SMD

0.33 (0.21–0.44)

 > 0

Yes

Sample restriction to high-quality trials

 Linde (1997) [6]

26

7

Random effects

OR

1.66 (1.33–2.08)

 > 1

Yes

 Linde (1997 and 1999) [6, 30]

26

7

Meta regression

OR

1.72 (1.28–2.31)

 > 1

Yes

 Shang (2005)/Lüdtke (2008) [32]

21

3

Random effects

OR

0.76 (0.59–0.99)

 < 1

Yes

 Mathie (2014) [10]

3

7

Random effects

OR

1.98 (1.16–3.38)

 > 1

Yes

 Mathie (2017) [11]

3

7

Random effects

SMD

0.18 (− 0.09 to + 0.46)

 > 0

No